基于贝尔图像亮度的新插值方法
首发时间:2008-11-06
摘要:大多数码相机通过彩色滤镜阵列获取图像,即贝尔图像。每个像素只能得到一个颜色值,而另外两种颜色值需要插值得到。基于两个重要的发现:1)通过贝尔图像频谱分析得到其亮度;2)图像在色度域内具有更平滑的特性,本文提出了基于贝尔图像亮度的新插值方法。此方法先利用滤波器模板提取贝尔图像亮度,然后将贝尔图像和亮度图像相减得出贝尔色度图像,在图像色度空域上利用双线性、平滑色相和基于色差的插值,最终和亮度图像相加得出全彩图像。本方法经验证不论在峰值信噪比上还是在视觉上都优秀与其他的方法。此方法计算简单,效果可靠,是一种简单优秀的插值方法。
For information in English, please click here
New Interpolation Method Based on the Luminance of the Bayer Pattern Image
Abstract:Most digital cameras acquire imagery with a color filter array(CFA), sampling only one color value for each pixel and interpolating the other two color values. We discuss in this paper two key observation : 1)obtain the luminance through the frequency analysis of a CFA image; 2) it is more smoother in image chrominance domain,and propose new interpolation methods. Our methods first extract the luminance using a filter, and then gain the chrominance in bayer format by subatracting the luminance image fron the CFA bayer image and interpolate in the chrominance domain using bilinear interpolation, smooth hue interpolation, and color difference interpolation , finally obtain the full color image. Experimental results on simulated CFA images verify that the proposed methods outperforms the other methods both visully and in terms of peak signal-to-noise ratio, at a notably lower computional cost. So our methods are simple and excellent.
Keywords: image interpolation color filter array(CFA) bayer pattern frequency analysis
基金:
论文图表:
引用
No.2549634617712259****
同行评议
共计0人参与
勘误表
基于贝尔图像亮度的新插值方法
评论
全部评论0/1000